The magnitude of the health and safety problems associated with alcohol consumption in the United States has resulted in Congress' mandating that a warning label appear on all containers of alcoholic beverages sold in this country. The goal of the proposed research is to define optimum designs for onproduct alcoholic beverage warnings by assessing the relative effectiveness of the mandated warning and a number of alternatives. The alternatives will vary in terms of wanting design criteria that have been shown to influence effectiveness. Specifically, design variables to be manipulated and studied will include: emphasis factors such as location, contrast, icons and borders; explicitness of hazard and consequences information, use of first versus third person text, and the presence and content of pictorials. The framework for the research can be characterized as a cognitive-action (or information processing) model. A number of effectiveness criteria will be employed that may be regarded as mediating variables in the continuum of cognitions and actions that ultimately affect morbidity and mortality rates. These criteria include attention demands, knowledge, attitudes, beliefs, decision outcomes and behavioral intentions. The questions to be answered are how do the different alcohol warning design parameters influence the effectiveness of the warning in attracting attention, in enhancing knowledge, in modifying attitudes and beliefs, and in effecting appropriate decisions and intended behavior. The methodology will consist of a series of studies (experiments) that will examine the effects of alternative warnings on these various measures. Generally, the warnings will be presented in actual contexts; that is, on the labels of alcoholic beverages. Attention effects will be assessed by measuring response times required to notice warnings and by using eye movement measurement techniques to analyze visual scanning patterns when examining alcohol container labels. Effects on knowledge, attitudes, and beliefs will be measured before and after exposure to the warnings by employing assessment instruments based on a great deal of previous work in this field. Scenarios (simulated circumstances) will be developed and used to determine relative warnings effects on decision outcomes and behavioral intentions. Long-term effects will be assessed by obtaining follow-up measures from subjects during a 6 month to 1 year period after their participation in the experiments. Studies will be carried out using subjects randomly selected to be representative of populations. Samples will include equal numbers from different ethnic groups [Blacks, Hispanics and Whites (not Hispanic)], men and women, and different age groups.